First record and impact of the crown-of-thorns starfish, Acanthaster planci (Spinulosida: Acanthasteridae) on corals of Malpelo Island, Colombian Pacific Krupskaya Narváez1 & Fernando A. Zapata2 1. Fundación Malpelo y Otros Ecosistemas Marinos. Carrera 11 No. 87-51 Local 4, Bogotá D.C., Colombia; [email protected] 2. Departamento de Biología, Universidad del Valle, Apartado Aéreo 25360, Cali, Colombia; [email protected] Received 30-VIII-2009. Corrected 11-XI-2009. Accepted 18-XII-2009. Abstract: The crown-of-thorns starfish, Acanthaster planci, is a major coral predator widely distributed in the Indo-Pacific region, where population outbreaks have caused dramatic impacts on coral reefs. In the Tropical Eastern Pacific (TEP) A. planci occurs at low population densities; it has been significantly abundant only in Panama and Cocos Island. We have recently found two individuals of A. planci at Malpelo Island, a small oce- anic island with significant coral communities located off the Colombian Pacific coast. The recent discovery of A. planci at Malpelo is significant in light of recent reports of its increased frequency of observation at Cocos Island and occurrence at the Galapagos Islands. The individuals found at Malpelo have been repeatedly sighted since 2004, actively feeding on nine species of coral. Although densities of A. planci are low at Malpelo and other oceanic islands of the TEP, these islands may act as stepping stones for the colonization of other coral reef areas in the region. However, the low densities of A. planci suggest that it currently poses no threat to coral reefs in the TEP. Rev. Biol. Trop. 58 (Suppl. 1): 139-143. Epub 2010 May 01. Key words: Acanthaster planci, coral reefs, crown-of-thorns starfish, Malpelo Island, Colombia, Tropical Eastern Pacific. The crown-of-thorns starfish Acanthaster in the Gulf of California, the Revillagigedo planci (Linnaeus, 1758) is a major preda- Islands, Cocos Island, mainland Costa Rica, tor of corals, and although a normal member and the Gulf of Chiriquí, Panamá (Porter 1972, of coral communities, widespread population Glynn 1974, Bakus 1975). Since then A. planci explosions have caused dramatic reductions in has also been reported for Isla del Caño, Costa coral cover on Indo-Pacific coral reefs (Endean Rica (Guzmán 1988), Clipperton Atoll (Glynn 1982). The causes of such explosions are et al. 1996) and Galápagos Islands (Hickman still not well understood (e.g., Pratchett 2005, 1998). A. planci has been generally considered Brodie et al. 2005) and because of their poten- to be absent from the Gulf of Panamá and the tially catastrophic effects, the occurrence of A. Colombian Pacific (e.g., Glynn 1974, Glynn & planci in coral reef areas where it was previ- Wellington 1983, Guzmán 1988, Reyes-Bonilla ously absent is significant. High densities of & Calderón-Aguilera 1999), but most informa- A. planci can also affect the relative abundance tion available for Colombia is derived from and diversity of coral and associated plant and studies carried out at Gorgona, a continental animal communities (Carpenter 1997). island, where A. planci has never been observed Before the mid-1970’s A. planci was known on coral reefs (Glynn et al. 1982, Zapata & to occur in the Tropical Eastern Pacific (TEP) Vargas-Ángel 2003). There is, however, one Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 58 (Suppl. 1): 139-143, May 2010 139 photographic record from the mid-1980’s at by one of us (K.N.) at Malpelo on August 8, a rocky outcrop on the Southernmost tip of 2004 at El Arrecife, a 2.4-ha coral community Gorgona (J.R. Cantera, pers. comm.). Although located on the Eastern side of the island. A A. planci was not observed at Malpelo Island single individual of 27cm in disc diameter with in 1972 (Birkeland et al. 1975, Downey 1975), 17arms was observed at 19m depth, where the it was subsequently observed there by sport temperature was 28ºC. The aboral margins of divers in 1991 at the site known as La Catedral the disc and arms were red, while their inner (J. Vélez, pers. comm.), and since 1998 at the portions were purple-gray or pink, and the cen- site known as La Nevera (S. Bessudo, pers. tral portion of the disc had a 4cm-wide, circular comm.). However, no formal report of these red belt surrounding the central gray periproct sightings were ever made. Here we report the with a single dark-red anus <1cm in diameter. occurrence of A. planci and comment on its The spines were 1-2cm long, purple-red or pink impact on corals at Malpelo Island based on and distributed throughout the aboral surface observations made since 2004. (Fig. 1). This individual was over a colony of Malpelo Island (4°00’05”N, 81°36’30”W) Pavona varians with its stomach everted and is a 1.6km-long rocky, oceanic island with sig- digesting the coral tissue. The same individual nificant coral communities and located approx- (based on size, number of arms and color pat- imately 500km West of the Colombian Pacific tern; see Glynn 1982) has been repeatedly seen coast (Birkeland et al. 1975, Zapata & Vargas- at different sites in El Arrecife during >40% of Ángel 2003). A. planci was first observed ~80 dives made between August 2004 and June Fig. 1. Acanthaster planci: Individual observed in several occasions at El Arrecife, Malpelo Island, Colombia. Photograph taken on June 22nd 2005, while the starfish was feeding on a colony of Pocillopora eydouxi. 140 Rev. Biol. Trop. (Int. J. Trop. Biol. ISSN-0034-7744) Vol. 58 (Suppl. 1): 139-143, May 2010 2006. A second individual of A. planci (22cm in were necessary for reef degradation (Endean disc diameter and 14 arms) was first observed 1977). It has been frequently stated that A. at La Nevera, a 0.5-ha coral community located planci has never reached outbreaking levels in on the Western side of Malpelo, on 9 February the TEP (Glynn 1974, Cortés 1997, Maté 2003) 2005, at 15m depth where the temperature was even though densities >15 ind./ha occurred in 22ºC. This individual was first found with its Panama during 1976-1979 (Glynn 1985, 1990) stomach everted feeding on a colony of Porites and at Cocos Is. in 1973 and 1987 (Glynn 1974, lobata and was seen during 5 of 79 dives made Guzmán & Cortés 1992). However, the absence between February and October 2005. of devastating effects of Acanthaster in Pana- Around the sites where both A. planci indi- ma has been attributed to the presence within viduals have been seen we usually found sev- colonies of the dominant reef building corals eral coral colonies with evident signs of recent (Pocillopora spp.) of crustacean symbionts that partial mortality, such as extensive white areas protect coral colonies, and a harlequin shrimp devoid of tissue or exposed skeletons begin- and a worm that prey on Acanthaster (Glynn ning to be colonized by algae. Presumably, this 2004). All other populations of A. planci in damage was caused by A. planci predation. the TEP have low densities (Reyes-Bonilla & We directly observed A. planci preying on Calderón-Aguilera 1999). At Malpelo, based nine species of coral (Pocillopora capitata, P. on estimates of coral reef areas at El Arrecife eydouxi, P. damicornis, Porites lobata, Pavona and La Nevera, one individual per reef repre- chiriquiensis, P. clavus, P. varians, P. gigantea sents 0.4 and 2.0 ind./ha, respectively. and Gardineroseris planulata). In general, the The recent increased frequency of obser- affected colonies were relatively large and vation of A. planci at Cocos Island ����������(N. Ghers- were not consumed totally by the starfish; most inichen, pers. comm.), and its appearance had >25% of living tissue remaining. By far where it was previously absent (Galápagos the most frequently consumed corals were P. and Malpelo) suggests a gradual, albeit slow, lobata and P. eydouxi in agreement with pre- increase of its geographic distribution in the vious observations at Cocos Is. (Guzmán & TEP. Oceanic islands could act as stepping Cortés 1992). However, both of these corals stones for further colonization in the region, but are abundant at Malpelo (Birkeland et al. 1975, the low densities of A. planci on these islands Garzón-Ferreira & Pinzón 1999) and there is suggest that these populations may not be sig- no indication that A. planci is selectively feed- nificant sources of propagules. Furthermore, ing on particular coral species as it has been the densities of A. planci in Panama have shown a case in Panama (Glynn 1974). decreasing trend in recent years (Fong & Glynn Population densities of A. planci have 1998, Glynn 2004) and densities elsewhere generally been low in the TEP, except in the in the TEP remain low. Although currently A. Gulf of Chiriquí, Panama, and Cocos Is., Costa planci does not pose a threat to coral reefs in Rica, where densities as high as 28 ind./ha have the TEP, evidence of further geographic spread been reported (Glynn 1974, 1990, Guzmán & or increased abundance in the region should Cortés 1992, Cortés 1997; the value of 250 receive attention. ind./ha listed by Reyes-Bonilla & Calderón- Aguilera [1999] represents an outlier based ACKNOWLEDGMENTS on an extrapolation of an aggregation of 25 ind./0.1ha reported by Glynn [1974]; P.W. We thank the Unidad Administrativa Espe- Glynn, pers. comm.). On the Great Barrier cial del Sistema de Parques Nacionales Natu- Reef, Australia, outbreaking populations of A. rales for permitting access to the Malpelo planci exhibit densities >15 ind./ha (Moran Island Sanctuary of Flora and Fauna, and N. & De’ath 1992), although earlier studies sug- Ghersinichen, J. Vélez, S.